The invention relates to a method for the analysis of essentially periodic signal sequences; in particular, physiological signals, such as EKG or the like, as well as relating to an apparatus for carrying out this method.
In the analysis of essentially periodic signal sequences, generally significant signal events (or occurrences) are employed for the purpose of generating trigger signals. The latter are then analyzed with regard to period, or frequency, respectively, of the significant events. In addition, the occurrence of extraordinary events in the periodic sequence of normal events is also generally of interest, since the latter influence, in particular, directly, the frequency measurement (or test) result.
Particularly in the analysis of physiological signal waveforms; for example, an electrocardiogram, in addition to the frequency measurement on the basis of the significant signal events, also the actual structure analysis of the signals receives general consideration. From the chronological progression of the occurrence of extrasystoles, for example, the medical practitioner can derive diagnostic information. Therefore, in the state of the art, it is customary to monitor the interval between the normal QRS-complexes for extrasystoles. Known electromedical apparatus operate with constantly specifiable (or predeterminable) time intervals in which the occurrence of extrasystoles in the EKG is detected and classified.
Information can also be derived from the omission of a normal event in an essentially periodic signal sequence. Specifically, in the EKG-analysis technology, in the case of lack of a normal QRS-complex, one speaks of a so-called asystolia, or "missed beat". In the case of EKG-recording on the viewing screen of an oscilloscope or on the strip chart of a cardiograh, such asystolias or "missed beats" can indeed be immediately recognized by the medical practitioner as anomalies; however, in the case of automatic analysis of EKG-curves, they are not readily recognized and evaluated as such.